Selfish Mining Fallacy - Debate III

"In this section, we argue that the current Bitcoin protocol has no measures to guarantee a low γ."

Well, actually, it does. The system already maintains a low gamma and they err in what impact a Sybil node has - again, they did not model the network topography. In fact, when I mention this, I am told that the network is "not assumed in the paper". No, it is not assumed, it is not tested and it is unknown. The authors just do not bother with empirical data or rigor. They assume it is different to what it is and that it must be fragile.

But why test the system? It is easier to spread doubt.

PS. I believe that people in here stated this as well...

"The random peer-to-peer structure of the Bitcoin overlay network will eventually propagate X to all miners, but the propagation of X under these conditions will be strictly slower than that of block P"

See, here is the thing... The authors DID assume a network structure.... Just the wrong one.

"Recall that these events occur at exponential intervals with an average frequency of α and (1−α), respectively".

Ummm. No. Wrong...

If α=1/3, (1-α)=2/3

The related time for α is:
E(α) = 10 min/α = 30 mins
E(1-α) = 10 min/(1-α) = 15 mins

Just a minor error...

The system state would even be ok... if it was independent... A shame it is not.

What the authors of the paper have done is to use the simplified model:
https://www.cs.utexas.edu/users/browne/cs380ns2003/Papers/SimpleQueuingModelspdf.pdf

In this simple model, we have independent events. Yes, hash rate is not dependent, but the strategy is. I know this seems count-intuitive and difficult, but when you have conditionals, you cannot treat them as being independent.

Where we have competing queues, we have a different model.
"Two Competing Queues with Linear Costs and Geometric Service Requirements: The μ cRule Is Often Optimal"
Author(s): J. S. Baras, A. J. Dorsey and A. M. Makowski
Source: Advances in Applied Probability, Vol. 17, No. 1 (Mar., 1985), pp. 186-209
Published by: Applied Probability Trust
Stable URL: http://www.jstor.org/stable/1427059
Accessed: 15-05-2017 19:43 UTC

I guess that they also do not understand that two events CAN occur in rapid succession. It is random. You cannot simply assume it is negligible and ignore it. That is not how probability calculations are made.

That means that a single hop state diagram is wrong.

Miners will mine for blocks and can find them in succession. It is infrequent, but needs to be a part of a valid model.

The system is modelled as a discrete-time process where it needs to be a continuous-time process.